The overall objective is to effect the total synthesis of trichothecene analogs related to trichodermol and verrucarol, compounds recognised as some of the most potent cytostatic substances known. A number of the naturally occurring compounds show potent in vivo activity against P-388 leukemia in mice, and in vitro activity against cells derived from human carcinoma of the nasopharynx. One member of this class of compounds, diacetoxyscirpenol (also known as anguidine), which is a constituent of "yellow rain" currently being employed by the Soviet Union as a chemical weapon, has already undergone extensive testing in the United States for its cancer-fighting potential. Most of the naturally occurring substances are highly toxic and are therefore of little therapeutic value. It is essential, then, to develop methods of total synthesis which allow the construction of structurally modified analogs, since the molecules are usually too sensitive to allow modification of the natural products. Unique methodology developed in our laboratories, using organometallic reagents, has already allowed us to synthesise one trichothecene analog, and we are now in a position to extend this approach to the preparation or highly oxygenated derivatives of potential therapeutic value. We propose to use tricarbonyl (4-methoxy-1-methyl cyclohexadienyl) iron hexafluoro-phosphate as a ring A precursor of trichodermol and its analogs in a unique coupling reaction with methyl potassio-2-oxocyclopentane carboxylate. This effects the formation of both carbocyclic rings of the target molecule and the formation of the two contiguous quarternary centres at an early stage. We plan to elaborate both diastereoisomers formed in this reaction to a single intermediate which can be diverted along several pathways to give trichodermol and synthetic analogs. Similar methodology for the synthesis of anguidine and verrucarol will be investigated.